Is Coolermaster making things up?

[tp4tissue]

1000Hz/1ms response time in USB mode    )

Without special USB drivers, how is this even possible.

The keyboard certainly doesn't come with special drivers like the Razer mices for 1000hz operation

When you plug in the quickfire rapid, it's just installed as a stock windows keyboard.

EDIT:
The Coolermaster guru, has informed us that the Quickfire Rapid is indeed a 125hz device, and not running at 1000hz natively.

He claims it was a publishing mistake.


If anyone would like to patch their, filco tkl, or coolermaster quickfire RAPID to 1000hz, PM me for details

[Soarer]

IIRC, the Pro is full-speed USB. So 1000Hz is no problem. No special drivers needed.

But for the low-speed USB ones, I think they are making it up.

I could search and check my facts... but so could you

[karljs]

It's not a perfectly definitive source, but it looks like the USB protocol does indeed support a poll rate up to 1000Hz:

http://www.codinghorror.com/blog/2007/04/mouse-dpi-and-usb-polling-rate.html

I would imagine that they don't work at that rate by default, but expect you to change it in software.

[tp4tissue]

It's not a perfectly definitive source, but it looks like the USB protocol does indeed support a poll rate up to 1000Hz:

http://www.codinghorror.com/blog/2007/04/mouse-dpi-and-usb-polling-rate.html

I would imagine that they don't work at that rate by default, but expect you to change it in software.

Yea that's exactly what I'm talking about.

I have my mouse setup with the custom usb drivers for 1000hz operation, and also used the same drivers for my keyboard.

there's however no way to definitively tell if the keyboard is being polled at 1000hz.

[Carter]

We dont make things up.

[tp4tissue]

We dont make things up.

So, exactly how is this board being polled at 1000hz with no custom drivers.??

Don't give me BS if you don't know. just leave..

[mariusmssj]

i got a quickfire pro keyboard which has the poll settings


Every time you choose a different setting, Windows loses the keyboard and says new device found and installs the new drivers, so my guess is that the keyboard had some kind of firmware built into it.

[shattentor]

Several mice like the Razer Abyssus have hardware switches under the mouse to adjust the polling rate without any drivers at all so I see no reason why keyboards shouldn't support >125Hz polling without drivers. I suppose it has something to do with the firmware, and USB is by its nature a universal bus that communicates with the device back and forth for identification purposes.

Here is also a proof that the Razer Deathadder operates on 1000Hz without drivers installed.

And BTW: >125Hz polling rates on keyboards are superfluous anyway 

[Soarer]

tp4 is just back to his neanderthalic trolling. He's still a believer in PS/2.

Unfortunately he's right about the Rapid (in an idiot savant kind of way; he gives completely the wrong reasoning).
Since it only uses low-speed USB, it cannot do 1000Hz (on Windows, without hacked drivers).

i got a quickfire pro keyboard which has the poll settings

I can't see the point in being able to change the setting! It's absolutely fine for a keyboard to operate at 1mS all the time. So I see that as a gimmick. Otherwise though, the Pro is one of the very few full-speed USB keyboards out there, which is a very welcome step forwards.

Several mice like the Razer Abyssus have hardware switches under the mouse to adjust the polling rate without any drivers at all so I see no reason why keyboards shouldn't support >125Hz polling without drivers. I suppose it has something to do with the firmware, and USB is by its nature a universal bus that communicates with the device back and forth for identification purposes.

Here is also a proof that the Razer Deathadder operates on 1000Hz without drivers installed.

And BTW: <125Hz polling rates on keyboards are superfluous anyway 

That would imply that 1000Hz Razers use full-speed USB then.
Which makes a lot more sense than relying on hacking the OS drivers!

I think you mean '>125Hz'
Sure there's a point to that - a few mS less delay. And no downsides.

[Lenfried]

Why do you cll him a idiot?

[Soarer]

I didn't!

[tp4tissue]

I didn't!

Oh! but you wanted to...

You're still holding a grudge about how I whooped ya back in the day, I see.  )

[Soarer]

No grudge... you whoop nothing but your own reputation as a 'minor authority' 

Rollbacks destroyed a lot of that thread That I do begrudge, since there was good info there, and I just can't be bothered to post it all again.

[tp4tissue]

No grudge... you whoop nothing but your own reputation as a 'minor authority' 

Rollbacks destroyed a lot of that thread That I do begrudge, since there was good info there, and I just can't be bothered to post it all again.

Oh please, anyone who's read that thread knows I won the internet. 

Besides, half of it was you guys steering the conversation into realms of which you have no understanding.

Circular logic is lost on you. 

[Carter]

I do apologize, I thought you were only referring to the Pro. The Rapid does not operate at 1000hz. Rapid is USB 1.1 so is actually 125hz, I have no idea how this info was published and will be removed immediately.

[tp4tissue]

I do apologize, I thought you were only referring to the Pro. The Rapid does not operate at 1000hz. Rapid is USB 1.1 so is actually 125hz, I have no idea how this info was published and will be removed immediately.

Thank-you Carter.

I only have one more question.

When I plug in the quickfire 'rapid

Device manager lists 2x USB Input devices, 1x Hid-compliant devices, 1x HID-compliant consumer control device, and 1x HID-keyboard device. This is all just the keyboard.

Why is this, and if I were to apply the USB 1000hz patch, do I apply it to all of them?

[alaricljs]

To do media keys 'properly' you register a 2nd device for the media stuff.  I don't know why that is.

[Soarer]

I do apologize, I thought you were only referring to the Pro. The Rapid does not operate at 1000hz. Rapid is USB 1.1 so is actually 125hz, I have no idea how this info was published and will be removed immediately.

We couldn't work out why either!

Device manager lists 2x USB Input devices, 1x Hid-compliant devices, 1x HID-compliant consumer control device, and 1x HID-keyboard device. This is all just the keyboard.

Why is this, and if I were to apply the USB 1000hz patch, do I apply it to all of them?

Power, multimedia controls, and keyboard.

AFAIK, the patch affects all low-speed USB devices in a blanket fashion. That's noted as one of its drawbacks.

[tp4tissue]

I do apologize, I thought you were only referring to the Pro. The Rapid does not operate at 1000hz. Rapid is USB 1.1 so is actually 125hz, I have no idea how this info was published and will be removed immediately.

We couldn't work out why either!

Device manager lists 2x USB Input devices, 1x Hid-compliant devices, 1x HID-compliant consumer control device, and 1x HID-keyboard device. This is all just the keyboard.

Why is this, and if I were to apply the USB 1000hz patch, do I apply it to all of them?

Power, multimedia controls, and keyboard.

AFAIK, the patch affects all low-speed USB devices in a blanket fashion. That's noted as one of its drawbacks.

Soarer, we should bury the hatchet. I honestly don't hate you as much as I used to.

Also, the "blanket fashion" patch was only under windows XP. In windows 7, the poll rate patch has to be done separately for each device.

Ontop of that, the windows 7 1000hz patch has quirks in it such that you can only patch one device at a time.


If anyone is looking to patch their coolermaster quickfire to 1000hz, PM me. it's very tricky.

[Soarer]

Also, the "blanket fashion" patch was only under windows XP. In windows 7, the poll rate patch has to be done separately for each device.

Ontop of that, the windows 7 1000hz patch has quirks in it such that you can only patch one device at a time.

Oh, I didn't know that! All I'd heard about Win7 was that it was a pain to patch. Neither Windows 7 or patching appeals to me; I'm a full-speed fan

So, just patch the keyboard device then. You don't need 1000Hz on your power and multimedia keys!

[Lanx]

could someone give me a quickie "idiots guide to full speed vs. low speed" this is a subject i've never looked into but wanted to, cuz everytime i plug my phone in (cuz of my charge station) i always get the windows "this device could be run faster..." thanks!

[tp4tissue]

could someone give me a quickie "idiots guide to full speed vs. low speed" this is a subject i've never looked into but wanted to, cuz everytime i plug my phone in (cuz of my charge station) i always get the windows "this device could be run faster..." thanks!

Um. If you are getting that windows message it means that the PORT that your device is plugged into is USB 1.0 or 1.1 limited, while the device itself supports USB 2.0

This only happens on older model computers where the front' connectivity USB ports are 1.0/1.1

Check to see if the USB ports on the BACK of your computer are USB 2.0  This is usually the case.

If your computer ONLY has 1.0/1.1 in Front and Back

Then, you can purchase a PCI or PCI'express USB expansion card to use USB 2.0

DO NOT buy a usb 3.0 expansion card, the current budget cheap offerings are total junk. The only good ones start at > $60, which is a total waste of money.

[Cart3r]

I do apologize, I thought you were only referring to the Pro. The Rapid does not operate at 1000hz. Rapid is USB 1.1 so is actually 125hz, I have no idea how this info was published and will be removed immediately.

The info is everywhere, on all websites, on the box. How could that happen?

[Soarer]

I do apologize, I thought you were only referring to the Pro. The Rapid does not operate at 1000hz. Rapid is USB 1.1 so is actually 125hz, I have no idea how this info was published and will be removed immediately.

The info is everywhere, on all websites, on the box. How could that happen?

You two talking together could get confusing!

In the endpoint descriptor of the Rapid, the bInterval field is set to 1. So my guess would be that an engineer mistakenly thought that that would result in 1mS (i.e. 1kHz), and somehow that got passed on to marketing without being checked.

It's actually a fairly understandable mistake by the engineer, since the USB 2.0 spec isn't 100% clear on the valid settings for bInterval.

In section 9.6.6 'Endpoint', it says:

bInterval - For full-/low-speed interrupt endpoints, the value of this field may be from 1 to 255.

That is in a section that pretty much anyone coding USB stuff will have to read early on.

But in section 5.7.4 'Interrupt Transfer Bus Access Constraints', it says:

An endpoint for an interrupt pipe specifies its desired bus access period. A full-speed endpoint can specify a desired period from 1 ms to 255 ms. Low-speed endpoints are limited to specifying only 10 ms to 255 ms.

Note that it doesn't use the identifier 'bInterval' there. It is used just after that, but at that point it is talking about high-speed, so at a glance it doesn't appear relevant.

Then in Windows, that 10mS will be mapped to 8mS, giving the well-known 125Hz. (link)

Both of those are harder to find. You pretty much have to be specifically looking for reasons why low-speed USB won't go faster than 125Hz (without driver hacks).

That may not be exactly how it happened... but it's plausible

I know, that isn't quite what you meant by how, but guessing anything further really would be pure speculation!

[Cart3r]

You two talking together could get confusing!

During the registration I wondered which person used my name. Now I know.

[Soarer]

You two talking together could get confusing!

During the registration I wondered which person used my name. Now I know.

Well, I checked your posts anyway - I don't think you're a nefarious sockpuppet!

[WhiteFireDragon]

It's hard to believe that ONE engineer is responsible for this mistake that gets published everywhere. And if it really is this one engineer, then I really question what kind of standards you need to work at CM. For an engineer to confuse 1ms with 1khz is just unacceptable, because this simple concept is taught in high school physics.

I think it's more reasonable to see this as purposely mistaken info for the sake of marketing, until some cares enough to bring it up (like in this thread).

[damorgue]

MX500 at 1000hz is to this day the best mice I have ever tried. They don't make sensors like that anymore.

[tp4tissue]

MX500 at 1000hz is to this day the best mice I have ever tried. They don't make sensors like that anymore.

It's pretty decent. Does it also have angle snapping, or was the 500 PRE-snap.

[Terrorsathan]

I was reading my QFR box today before packing it away, and it defiantly says 1ms

[damorgue]

MX500 at 1000hz is to this day the best mice I have ever tried. They don't make sensors like that anymore.

It's pretty decent. Does it also have angle snapping, or was the 500 PRE-snap.

Pre-increase. They increased it a bit with 510 and 518. There is still some though, but I am not bothered with it. I think it is a larger problem in high sensitivity mice where it is more noticeable.

[Lanx]

actually a mod should rename
cart3r
to something else, the real
carter
is an actual rep from the industry, while the cart3r, is just someone
i mean it's like if some noob was going around calling himself eMav.

[Soarer]

It's hard to believe that ONE engineer is responsible for this mistake that gets published everywhere. And if it really is this one engineer, then I really question what kind of standards you need to work at CM. For an engineer to confuse 1ms with 1khz is just unacceptable, because this simple concept is taught in high school physics.

Err... a signal with a period of 1mS does have a frequency of 1kHz! And that is exactly what the bInterval setting does when set to 1 in full-speed mode. Did you mean to say something else?

I think it's more reasonable to see this as purposely mistaken info for the sake of marketing, until some cares enough to bring it up (like in this thread).

Sure, that's possible too. It could also just be confusion between the Pro and the Rapid, since the Pro does have 1kHz. And I detailed the engineer possibility above because that info would not be known to most people.

It's also possible that the supplier of the controller chips is 'responsible' somehow. Don't forget there have also been Filco models that claim "data transmission  to the maximum" as well, and they are also only low-speed USB.

[dorkvader]

It's hard to believe that ONE engineer is responsible for this mistake that gets published everywhere. And if it really is this one engineer, then I really question what kind of standards you need to work at CM. For an engineer to confuse 1ms with 1khz is just unacceptable, because this simple concept is taught in high school physics.

Err... a signal with a period of 1mS does have a frequency of 1kHz! And that is exactly what the bInterval setting does when set to 1 in full-speed mode. Did you mean to say something else?

I thought he meant something else. Take, for instance the difference between Hz and Becquerels, If your time between events is 1ms on average, but irregular, it's not Hz. Also, looking at PWM circuitry, if one considers events that take time, it might be able to scan the 'board at 1Khz, but it can only send data every few Ms. Or maybe it can scan the board is 1ms, but it takes a couple ms between scans to match up with the sending of data. I'm not aware of the nitty gritty of it all, but there are a couple of differences that would confuse someone between 1Khz and 1ms.

But you're right. If events happen regularly 1 ms apart, then that corresponds to a rate of 1 KHz.
If an event is 1 ms long, and they occur 9 ms apart, then you get a whole 'nother frequency (100 Hz).

[tp4tissue]

It's hard to believe that ONE engineer is responsible for this mistake that gets published everywhere. And if it really is this one engineer, then I really question what kind of standards you need to work at CM. For an engineer to confuse 1ms with 1khz is just unacceptable, because this simple concept is taught in high school physics.

Err... a signal with a period of 1mS does have a frequency of 1kHz! And that is exactly what the bInterval setting does when set to 1 in full-speed mode. Did you mean to say something else?

I thought he meant something else. Take, for instance the difference between Hz and Becquerels, If your time between events is 1ms on average, but irregular, it's not Hz. Also, looking at PWM circuitry, if one considers events that take time, it might be able to scan the 'board at 1Khz, but it can only send data every few Ms. Or maybe it can scan the board is 1ms, but it takes a couple ms between scans to match up with the sending of data. I'm not aware of the nitty gritty of it all, but there are a couple of differences that would confuse someone between 1Khz and 1ms.

But you're right. If events happen regularly 1 ms apart, then that corresponds to a rate of 1 KHz.
If an event is 1 ms long, and they occur 9 ms apart, then you get a whole 'nother frequency (100 Hz).

What we have here people is a clear cut case of the Geekhack brand, "Lost In Translation"

[Soarer]

It's hard to believe that ONE engineer is responsible for this mistake that gets published everywhere. And if it really is this one engineer, then I really question what kind of standards you need to work at CM. For an engineer to confuse 1ms with 1khz is just unacceptable, because this simple concept is taught in high school physics.

Err... a signal with a period of 1mS does have a frequency of 1kHz! And that is exactly what the bInterval setting does when set to 1 in full-speed mode. Did you mean to say something else?

I thought he meant something else. Take, for instance the difference between Hz and Becquerels, If your time between events is 1ms on average, but irregular, it's not Hz.

If it's mS on average, then it has some average Hz as well!

Also, looking at PWM circuitry, if one considers events that take time, it might be able to scan the 'board at 1Khz, but it can only send data every few Ms.

PWM frequency is given as the whole repeating cycle.

Or maybe it can scan the board is 1ms, but it takes a couple ms between scans to match up with the sending of data. I'm not aware of the nitty gritty of it all, but there are a couple of differences that would confuse someone between 1Khz and 1ms.

Yeah, that's possible. I think I pondered on that somehwere, not sure if it was in relation to the Rapid or the Filco. IIRC we came to the conclusion that both intended to mean the USB polling rate, and not the matrix scan rate.

The Rapid's specifications on the website still say "Polling Rate 1000 Hz /1 ms", which could mean either, but I think most people would understand it to mean USB polling. The same spec is given for the Pro. "Scanning rate" would be an appropriate change for clarification, if that is the case.

What does it say on the box?

But you're right. If events happen regularly 1 ms apart, then that corresponds to a rate of 1 KHz.
If an event is 1 ms long, and they occur 9 ms apart, then you get a whole 'nother frequency (100 Hz).

The second case there would have a period of 10mS

Certainly when talking about the bInteval field in the endpoint, it's straightforward